• Register
  • Help
Results 1 to 3 of 3

Topic: Level mismatch between soft synths and samples...

  1. #1

    Level mismatch between soft synths and samples...

    Here's a quandry.

    On my PC-based live performance system, I use Gigastudio running various emulations of real instruments - PMI Bosendorfer, Scarbee Rhodes, QL Brass etc - plus NI B4 and their Pro-53 for more synthy sounds.

    I always find that, with the various software gain stages set at average kinds of volumes, the synths come out WAY louder than the samples, like several order of magintude louder.

    If I set my rig up around the kind of levels coming out of Giga, then I need to transmit a MIDI cc7 value to set the synths on like under 20, out of the full 127, to avoid blasting the hell out of my whole system (and ears). Sometimes even lower than that, or I need to set cc11 really low as well.

    I don't suppose this really matters, except that I need to run my 300w keyboard up most of the way up to get decent level. It doesn't seem to make a lot of sense to squeeze the soft synth output so much, then have to amplify it so much at the analogue stage in compensation.

    Why is Giga's output level so low?

    Would the same issue still apply if I switched to running the samples in another host, such as Kontakt?

    Is there some way that I can substantially amplify the output signal from Giga while still in software, eg by sending it via some other app before it reaches my soundcard, without creating artifacts, so I can take average levels from the synths instead and take some of the heat off my amp?

    Thanks for any tips.

  2. #2

    Re: Level mismatch between soft synths and samples...

    I've never run into that dramatic a difference, but I have run into level problems between software synths and samplers from different developers.

    My solution, and it may seem a little odd, is to group signals that are too low together and buss them out to a separate analog output, and then mix them with the rest in the analog domain.


    Well, if you insert too much attenuation within a DAW or software synth you end up reducing the resolution, and ultimately the signal to noise ratio - neither of which is a good thing<G>!

    And if you apply too much gain you also reduce the resolution, and again ultimately you tend to decrease the signal to noise ratio, this time because you raise the noise floor.

    I use two computers in my current configuration. The first hosts Sonar and all of my DirectX or VST instruments. The second if dedicated to GigaStudio.

    For whatever reason, I do not run into significant level problems with this configuration, and I've been able to transfer the audio from GigaStudio to the main machine via a pair of lightpipe connections. Every once in a while I do run into a case where a specific library may be a bit on the quiet side, but thus far the difference has been on the order of 3 - 6 dB, and I am comfortable making up the difference in Sonar.

    On the main machine the differences between different developers also fall into that range, and while it is frustrating, I can make it work by tweaking the settings of the individual patches. It's tedious, time consuming, and really ought to be unneccesary, but there you go... it is necessary.

    If you are using only one machine, and the synths are louder than the sampler, I'd probably edit the synth patches as my first pass. If that doesn't work I'd convert all the outputs to audio, and then try to balance gain for the samples with attenuation for the synths to get to where I wanted to be.

    BTW, forgot to mention that among other things, I use Scarbee's bass libraries, the QL Brass, and NI's FM7, Pro-53, and B4. The NI synths are somewhat louder than GS, but for FM7 and Pro-53 I've found it fairly simply to tweak them into submission.

    I recently picked up the Gforce Vintage bundle, and they are well behaved. So too are the plugins that came with Sonar and Project 5.
    Last edited by wst3ae; 10-24-2006 at 11:26 AM. Reason: additional info
    Bill Thompson
    Audio Enterprise

  3. #3
    Senior Member Steve_Karl's Avatar
    Join Date
    May 2004
    Pittsburgh, PA 15206 USA

    Re: Level mismatch between soft synths and samples...

    I'm running Giga 2.5 with KH Emerald and many other samples from other libs.
    and also EWQL Gold, Stormdrum, and Absynth and all of my levels are very close to bing the same.

Go Back to forum


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts